Two interrelated aims guide our research with very-low-birthweight infants (VLBW: less than 1500 grams): (1) To describe how VLBW infants differing in perinatal biological risk differ also in their characteristics as early social partners and (2) to assess the impact of these differing characteristics upon the development of mother-infant interactions during the first 3 months at home. In Study 1 two groups of VLBW infants (high vs. lower biological risk) are assessed for the functional relationships existing between infant responsiveness and increasing amounts of social stimulation under conditions (prefeeding vs. postfeeding, swaddled vs. unswaddled) varying in their demands upon the infant for behavioral organization. These contrasts occur in the nursery at 32, 34, and 36 weeks' postconceptional ages and between the VLBW groups and a full-term comparison group at time of discharge from the hospital. Study 2 uses these findings to compose two VLBW groups differing in behavioral risk (more vs. less problematic social partner at discharge from hospital). These groups are contrasted in (1) behavior during mother-infant enface interactions at 2, 6, and 10 weeks after discharge; (2) maternal interview data about the experience of interacting with their infant; and (3) the infant's social response characteristics assessed independently of the mother at 4 months' postconceptional age. The impact of the infant's behavior upon the mother is assessed through summary measures of mother's activity levels and sensitivity during mother-infant interaction, mother's reported affects, attributions, and intentions during interaction, and sequential analyses of the impact of specific infant behaviors (e.g., gaze aversion). To determine how the infant's behavioral characteristics and the mother's reactions mediate between perinatal biological risk and subsequent development (1) changes in the infant's social response characteristics from discharge to 4 months are related to the mother's reactions during the intervening months of enface interaction and (2) path analysis techniques assess a developmental model linking biological risk, problematic infant behavioral characteristics and mother's active, insensitive behavior during enface interaction to subsequent problems of development.